The Resource is using molecular dynamics methods to study theinteraction of steroid hormone receptors with DNA. Receptors inthis class of proteins recognize and bind to specific sequences ofDNA and play a crucial role in a variety of biological processesincluding metabolism, stress response, and the development ofsecondary sexual characteristics. One member of this class, theestrogen receptor, is also used for the diagnosis and treatmentof certain types of breast cancer. The goal of simulations at theResource is to determine the effects of protein-DNA complex formation on the conformation of the interacting molecules and togain an understanding of the nature of sequence specific DNArecognition by proteins. Simulations of the complex of a dimer of estrogen receptor DNAbinding domains (ER-DBD), with DNA have been conductedfor two sequences of DNA: the palindromic consensus sequenceds(CAGGTCACAGTGACCTG) and the non-consensus, yet biologicallyactive, sequence ds(CAGAACACAGTGACCTG). Each system consisted of approximately 36; 000 atoms including Na+ ions for net chargeneutrality. We have utilized the Resource cluster and the program NAMD [5]to compute two 100 ps trajectories, one for each system.Eachsimulation required 20 days of parallel computation on eight HP workstations.These simulations represent the first productionruns of NAMD utilizing the Coulomb force evaluation packagedpmta [44] which allows one to efficiently determine electrostaticinteractions for large systems without cut-off.These prototype simulations led to the addition of several user features, the modification of several routines for improved performance and the discovery and correction of several programming errors in NAMD. A preliminary analysis of the trajectories indicates that the twobase pair substitutions altered the protein-DNA recognition and,thus, the global conformation of the complex. This is most readily observed as differences in the conformation of the DNA helical axis and differences in the relative orientation of the ER-DBD's between the two simulations. Complete analysis is being conducted to confirm and quantify these results.